{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-07-23T08:01:28.298Z","headline":"NVIDIA 发布 Nemotron 3 Embed 开源嵌入模型系列，旗舰版登顶 RTEB 榜首","description":"NVIDIA 发布开源嵌入模型系列 Nemotron 3 Embed，包含三个模型。旗舰版 Nemotron-3-Embed-8B-BF16 在 RTEB 排行榜上排名第一，得分 78.5%。","url":"https://www.aioga.com/news/cmrnqu5q801otbizzgvoejz74/","mainEntityOfPage":"https://www.aioga.com/news/cmrnqu5q801otbizzgvoejz74/","datePublished":"2026-07-16T16:01:21.000Z","dateModified":"2026-07-16T16:01:21.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://huggingface.co/blog/nvidia/nemotron-3-embed-wins-rteb","https://aihot.virxact.com/items/cmrnqu5q801otbizzgvoejz74"],"canonicalUrl":"https://www.aioga.com/news/cmrnqu5q801otbizzgvoejz74/","directAnswer":{"@type":"Answer","text":"Aioga 编辑摘要：NVIDIA 发布开源嵌入模型系列 Nemotron 3 Embed，包含三个模型。 Aioga 将其归入「AI资讯」方向，重点关注它对真实使用和行业竞争的影响。","url":"https://www.aioga.com/news/cmrnqu5q801otbizzgvoejz74/","dateCreated":"2026-07-16T16:01:21.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"huggingface.co source article","url":"https://huggingface.co/blog/nvidia/nemotron-3-embed-wins-rteb","datePublished":"2026-07-16T16:01:21.000Z","provider":{"@type":"Organization","name":"huggingface.co","url":"https://huggingface.co/blog/nvidia/nemotron-3-embed-wins-rteb"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.virxact.com/items/cmrnqu5q801otbizzgvoejz74","datePublished":"2026-07-16T16:01:21.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.virxact.com/items/cmrnqu5q801otbizzgvoejz74"}}],"aggregationSource":"Hugging Face：Blog（RSS）","originalPublisher":{"name":"huggingface.co","url":"https://huggingface.co/blog/nvidia/nemotron-3-embed-wins-rteb"},"article":{"id":"cmrnqu5q801otbizzgvoejz74","slug":"cmrnqu5q801otbizzgvoejz74","url":"https://www.aioga.com/news/cmrnqu5q801otbizzgvoejz74/","title":"NVIDIA 发布 Nemotron 3 Embed 开源嵌入模型系列，旗舰版登顶 RTEB 榜首","title_en":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"NVIDIA 发布开源嵌入模型系列 Nemotron 3 Embed，包含三个模型。旗舰版 Nemotron-3-Embed-8B-BF16 在 RTEB 排行榜上排名第一，得分 78.5%。","source":"Hugging Face：Blog（RSS）","sourceUrl":"https://huggingface.co/blog/nvidia/nemotron-3-embed-wins-rteb","aiHotUrl":"https://aihot.virxact.com/items/cmrnqu5q801otbizzgvoejz74","publishedAt":"2026-07-16T16:01:21.000Z","category":"AI资讯","score":0,"selected":false,"articleBody":["Evaluation: Retrieval Quality, Agentic Efficiency, and Deployment Tradeoffs ：#evaluation-retrieval-quality-agentic-efficiency-and-deployment-tradeoffs RTEB Leadership and Strong Gains Across Retrieval Benchmarks ：#rteb-leadership-and-strong-gains-across-retrieval-benchmarks Why Better Retrieval Matters for Agents ：#why-better-retrieval-matters-for-agents Scaling Retrieval with NVFP4 on Blackwell ：#scaling-retrieval-with-nvfp4-on-blackwell Day 0 Performant NIM ：#day-0-performant-nim How We Built the Nemotron 3 Embed Models ：#how-we-built-the-nemotron-3-embed-models Scaling Down to 1B ：#scaling-down-to-1b Enterprise Partner Evaluations ：#enterprise-partner-evaluations Getting Started ：#getting-started Retrieval is critical in multi-step agentic workflows where poor retrieval can cause agents to fetch irrelevant context, re-query, waste token budget, and carry noise into later reasoning steps.","Today, we are releasing NVIDIA Nemotron 3 Embed ：https://huggingface.co/collections/nvidia/nemotron-3-embed, a collection of open and commercially available embedding models designed to improve retrieval quality while giving developers practical deployment options for production-scale RAG, agentic retrieval, code retrieval, and agent memory.","The collection includes three open models that achieve state-of-the-art retrieval across the accuracy-efficiency curve, led by an 8B model that tops the RTEB leaderboard and efficient 1B variants built for production-scale deployment:","Table 1. Nemotron 3 Embed Model Usability and Deployment Matrix.","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/I-QbjjUGW9tmE2mcDaW2f.png","Figure 1. RTEB Multilingual Leaderboard：https://mteb-leaderboard.hf.space/benchmark/RTEB%28beta%29 screenshot (July 15, 2026) showing Nemotron-3-Embed-8B-BF16 ranked as #1.","Beyond the RTEB result, Nemotron 3 Embed introduces a production-ready feature set for enterprise retrieval deployments:","We evaluate Nemotron 3 Embed across three dimensions: retrieval quality, downstream agentic efficiency, and deployment tradeoffs. The 8B model establishes the model collection’s quality ceiling, while the 1B BF16 and NVFP4 variants bring the same retrieval-focused design to lower-cost and higher-throughput deployment settings.","We first evaluated the models on RTEB：https://huggingface.co/blog/rteb, where Nemotron-3-Embed-8B-BF16 ranks #1. We also tested these models across ViDoRe V3：https://huggingface.co/blog/QuentinJG/introducing-vidore-v3 Text, and MMTEB：https://arxiv.org/abs/2502.13595 Retrieval and LongEmbed：https://github.com/dwzhu-pku/longembed using average NDCG@10.","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/_4Rzuu2UMDD1DeHs-TOhP.png","Figure 2. Retrieval accuracy using average NDCG@10 across RTEB, ViDoRe V3 Text, MMTEB Retrieval and LongEmbed, comparing the Nemotron 3 Embed models with prior-generation Nemotron baselines.","To evaluate retrieval in an agentic setting, we use a search agent：https://huggingface.co/blog/nvidia/nemo-retriever-agentic-retrieval powered by Nemotron 3 Ultra：https://github.com/NVIDIA-NeMo/Nemotron/tree/main/usage-cookbook/Nemotron-3-Ultra and vary the embedding model used by the retrieval system. Better retrieval can return relevant evidence earlier, helping the agent avoid repeated searches, unnecessary reasoning turns, and extra context inspection. We compare average retrieval accuracy with estimated downstream agentic token cost per query across ViDoRe V3, BRIGHT：https://brightbenchmark.github.io/, and BrowseComp-Plus：https://huggingface.co/spaces/Tevatron/BrowseComp-Plus.","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/v45KGVKGsO_ZS8WW-KUgU.png","Figure 3. Average retrieval accuracy versus downstream agentic token cost per query across ViDoRe V3, BRIGHT, and BrowseComp-Plus.","Evaluation note: The search agent uses Nemotron 3 Ultra. Downstream token cost is estimated from Nemotron 3 Ultra input/output token counts using the GPT-5.5 pricing formula.","Figure 3 shows that stronger retrieval reduces downstream agentic token cost. More accurate retrievers return relevant evidence earlier, which helps agents complete tasks with fewer repeated searches and fewer reasoning turns. In these evaluations, the Nemotron 3 Embed models improve the agentic retrieval frontier, with the 8B model delivering both the highest average retrieval accuracy and the lowest estimated downstream token cost across ViDoRe V3, BRIGHT, and BrowseComp-Plus.","For high-throughput deployments, teams often choose smaller embedding models to meet latency and cost targets. Nemotron-3-Embed-1B-NVFP4 is designed to narrow the gap between serving efficiency and retrieval quality by using native NVFP4 acceleration on NVIDIA Blackwell architectures：https://www.nvidia.com/en-us/data-center/technologies/blackwell-architecture/. The model quantizes the weights and activations of linear layers to NVFP4 for efficient inference, and uses Quantization-Aware Distillation (QAD) to help recover accuracy for long input sequences.","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/R1QHlEFRytVBJcvt8lL9i.png","Figure 4. ViDoRe V3 retrieval accuracy versus serving efficiency, comparing Nemotron-3-Embed-1B-NVFP4 with selected smaller open embedding baselines, including Qwen3-Embedding-0.6B and EmbeddingGemma-300M.","For production-scale retrieval systems, the serving stack also needs to preserve that efficiency under real request loads, across different input sequence lengths and hardware targets. To make Nemotron 3 Embed performant at enterprise scale today, we are also releasing an optimized NVIDIA NIM microservice for the 1B model. As shown in Figure 5, the Rust-based Nemotron 3 Embed NIM matches or outperforms the vLLM checkpoint on NVIDIA GB200 and RTX PRO 6000 GPUs across ISLs of 256 & 1024.","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/ZFfiQHKsv2surfaCBUCkR.png","Figure 5. Nemotron 3 Embed NIM serving performance compared with the vLLM checkpoint on NVIDIA GB200 and RTX PRO 6000 GPUs.","Nemotron-3-Embed-8B-BF16 adapts the Ministral-3-8B-Instruct-2512：https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512 backbone by converting its causal decoder into a bidirectional encoder for full-sequence retrieval. The model is trained with contrastive pre-training on a blend of web-sourced and synthetic text pairs, then fine-tuned on curated multilingual retrieval datasets across domains such as legal, finance, medical, business, and education. This 8B model serves as the flagship embedding model, while earlier 8B teacher checkpoints from the same development line were used to distill the efficient 1B variants.","The 1B model is not a small retriever trained from scratch. We first applied the bidirectional adaptation recipe to the Ministral-3-3B-Instruct-2512：https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512 backbone to establish a 3B retriever base, then compressed it through two-rounds of structured pruning and distillation.","First, the 3B parent model was compressed to a 2B intermediate footprint using NVIDIA ModelOpt：https://github.com/NVIDIA/Model-Optimizer’s mcore_minitron Neural Architecture Search engine. The NAS pipeline searched across hidden width, FFN size, attention heads, and depth under a strict parameter budget to identify an efficient architecture for retrieval workloads.","The resulting 2B intermediate model was then distilled from an 8B teacher checkpoint to recover ranking accuracy. We used a combined cosine distance loss and mean squared error loss on a multilingual, in-domain retrieval data blend to align the student’s embeddings with the teacher.","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/U6hbjPbLo70KwwtbAs2np.png","Figure 6. The pruning and distillation pipeline compresses the retriever from a 3B base to the final 1B production model.","This same sequence, ModelOpt structured pruning followed by 8B teacher distillation, was repeated a second time to compress the 2B intermediate model down to the final 1.14B embedding model. Final training used a progressive two-stage context-scaling schedule:","The following table summarizes the core technical specifications and deployment targets for the Nemotron 3 Embed models:","Table 2. Architectural specifications and core inference configurations for the Nemotron 3 Embed models.","Enterprise ISVs, AI-native companies, data platform companies, and memory providers are already evaluating Nemotron 3 Embed across agentic retrieval, agent memory, code retrieval, and production inference workflows.","NVIDIA Nemotron 3 Embed is released with open weights and open-source training recipes, giving organizations full control over how retrieval models are customized and deployed for production AI applications.","Developers can get started using the deployment option that best fits their workflow:","For workloads requiring domain adaptation or footprint reduction, we have also open-sourced NVIDIA NeMo AutoModel：https://github.com/nvidia-nemo/automodel training recipes:","For example, on the NV Docs：https://huggingface.co/datasets/nvidia/Retrieval-Synthetic-NVDocs-v1 evaluation, fine-tuning Nemotron-3-Embed-1B-BF16 improved NDCG@10 from 56.7% to 63.3% (+11.6%) and Recall@5 from 56.1% to 62.8% (+11.9%).","Whether you are building enterprise search, production RAG, agent memory, code retrieval, or agentic AI systems, NVIDIA Nemotron 3 Embed provides flexible deployment options—from open-source models on Hugging Face to fully managed AI Cloud platforms and NVIDIA NIM microservices.","Explore the models, deploy them on your preferred platform, and let us know what you are building.","Fair and Disentangled Evaluation of Deep-Research Agents","Congratulations on the LMEB SOTA! 🎉","The results merged in https://github.com/embeddings-benchmark/results/pull/617：https://github.com/embeddings-benchmark/results/pull/617 show outstanding LMEB performance:","Nemotron-3-Embed-8B-BF16: 64.4, a new overall SOTA. Nemotron-3-Embed-1B-BF16: 61.5, a new SOTA at the ~1B scale.","The LMEB leaderboard：https://mteb-leaderboard.hf.space/benchmark/LMEB appears to be awaiting synchronization.","It's live now! Thank you for evaluating our models there!"],"articleImages":[{"sourceUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1594197321835-5f057efe5d08220171a0ad8f.jpeg","alt":"","afterParagraph":0,"url":"/media/articles/cmrnqu5q801otbizzgvoejz74/83f80501d0c501ea.webp"},{"sourceUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/1605114051380-noauth.jpeg","alt":"","afterParagraph":0,"url":"/media/articles/cmrnqu5q801otbizzgvoejz74/3ec7f542e3247cfe.webp"},{"sourceUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5fc181c4ea82dd667bb0ffae/FeiX5mrJakX7FpgK5zQli.png","alt":"","afterParagraph":0,"url":"/media/articles/cmrnqu5q801otbizzgvoejz74/f4d0e60381f33d79.webp"},{"sourceUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/5ff5943752c26e9bc240bada/Exyzf3C_gJ2KdsL4K5_cq.png","alt":"","afterParagraph":0,"url":"/media/articles/cmrnqu5q801otbizzgvoejz74/0d50bc08e6883650.webp"}],"mediaStatus":"ok","articleBodyZh":["评估：检索质量、代理效率与部署权衡：#evaluation-检索质量-代理效率与部署权衡 RTEB 在检索基准中的领先与强劲收益：#rteb-领导力与跨检索基准的强劲提升 为什么更好的检索对代理至关重要：#why-更好的检索对代理至关重要 在Blackwell上用NVFP4扩展检索：#scaling-retrieval-with-nvfp4-on-blackwell 第0天 表现NIM：#day-0-performant-nim我们如何构建Nemotron 3嵌入模型：#how-we-built-the-nemotron-3-embed-models 缩减至1B：#scaling-down-to-1b 企业合作伙伴评估：#enterprise-partner-evaluations 入门：#getting-started 检索在多步代理工作流中至关重要，因为检索不良可能导致代理获取无关上下文、重新查询、浪费令牌预算，并将噪声带入后续推理步骤。","今天，我们发布了NVIDIA Nemotron 3 Embed：https://huggingface.co/collections/nvidia/nemotron-3-embed，这是一套开放且商业化的嵌入模型集合，旨在提升检索质量，同时为开发者提供生产规模RAG、代理检索、代码检索和代理内存的实用部署选项。","该收藏包含三种开放模型，实现了准确率-效率曲线上的最先进检索，其中以 8B 型号（位居 RTEB 排行榜第一）和高效 1B 版本为首，专为生产规模部署而设计：","表1。Nemotron 3 嵌入模型可用性与部署矩阵。","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/I-QbjjUGW9tmE2mcDaW2f.png","图1。RTEB 多语言排行榜：https：//mteb-leaderboard.hf.space/benchmark/RTEB%28beta%29 截图（2026年7月15日），显示 Nemotron-3-Embed-8B-BF16 排名为 #1。","除了RTEB结果外，Nemotron 3 Embed还引入了适用于企业检索部署的生产级功能集：","我们从三个维度评估Nemotron 3 Embed：检索质量、下游代理效率和部署权衡。8B型号确立了模型系列的质量上限，而1B型BF16和NVFP4变体则将同样以回收为重点的设计带入了低成本、高通量的部署环境。","我们首先在 RTEB 上评估了这些模型：https://huggingface.co/blog/rteb，其中 Nemotron-3-Embed-8B-BF16 排名第一。我们还在 ViDoRe V3：https://huggingface.co/blog/QuentinJG/introducing-vidore-v3 文本，以及 MMTEB：https://arxiv.org/abs/2502.13595 检索和 LongEmbed：https://github.com/dwzhu-pku/longembed 使用平均 NDCG@10 测试了这些模型。","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/_4Rzuu2UMDD1DeHs-TOhP.png","图 2. 使用平均 NDCG@10 对 RTEB、ViDoRe V3 文本、MMTEB 检索和 LongEmbed 的检索准确性进行比较，将 Nemotron 3 Embed 模型与先前的 Nemotron 基线进行比较。","为了在具有代理性的设置中评估检索，我们使用一个搜索代理：https://huggingface.co/blog/nvidia/nemo-retriever-agentic-retrieval，由 Nemotron 3 Ultra 驱动：https://github.com/NVIDIA-NeMo/Nemotron/tree/main/usage-cookbook/Nemotron-3-Ultra，并改变检索系统使用的嵌入模型。更好的检索可以更早返回相关证据，帮助代理避免重复搜索、不必要的推理步骤和额外的上下文检查。我们在 ViDoRe V3、BRIGHT：https://brightbenchmark.github.io/ 和 BrowseComp-Plus：https://huggingface.co/spaces/Tevatron/BrowseComp-Plus 中比较了平均检索准确性与每次查询的估计下游代理令牌成本。","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/v45KGVKGsO_ZS8WW-KUgU.png","图 3. 在 ViDoRe V3、BRIGHT 和 BrowseComp-Plus 中，每次查询的下游代理令牌成本与平均检索准确性比较。","评估说明：搜索代理使用 Nemotron 3 Ultra。下游令牌成本根据 Nemotron 3 Ultra 输入/输出令牌数量，使用 GPT-5.5 定价公式估算。","图 3 显示，更强的检索可以降低下游代理令牌成本。更准确的检索器更早返回相关证据，有助于代理以更少的重复搜索和推理步骤完成任务。在这些评估中，Nemotron 3 Embed 模型改善了代理检索前沿，其中 8B 模型在 ViDoRe V3、BRIGHT 和 BrowseComp-Plus 中提供了最高的平均检索准确性和最低的估计下游令牌成本。","对于高吞吐量的部署，团队通常会选择较小的嵌入模型以满足延迟和成本目标。Nemotron-3-Embed-1B-NVFP4 旨在通过在 NVIDIA Blackwell 架构上使用原生 NVFP4 加速来缩小服务效率与检索质量之间的差距：https://www.nvidia.com/en-us/data-center/technologies/blackwell-architecture/。该模型将线性层的权重和激活量量化为 NVFP4 以实现高效推理，并使用量化感知蒸馏（Quantization-Aware Distillation, QAD）来帮助恢复长输入序列的精度。","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/R1QHlEFRytVBJcvt8lL9i.png","图 4. ViDoRe V3 检索精度与服务效率比较，将 Nemotron-3-Embed-1B-NVFP4 与选定的小型开源嵌入基线进行比较，包括 Qwen3-Embedding-0.6B 和 EmbeddingGemma-300M。","对于生产级别的检索系统，服务栈还需要在真实请求负载下保持效率，覆盖不同的输入序列长度和硬件目标。为了使 Nemotron 3 Embed 在企业规模下高效运行，我们还发布了针对 1B 模型优化的 NVIDIA NIM 微服务。如图 5 所示，基于 Rust 的 Nemotron 3 Embed NIM 在 NVIDIA GB200 和 RTX PRO 6000 GPU 上，在 ISL 为 256 和 1024 时，其表现与 vLLM 检查点相当或更优。","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/ZFfiQHKsv2surfaCBUCkR.png","图 5. Nemotron 3 Embed NIM 的服务性能与 NVIDIA GB200 和 RTX PRO 6000 GPU 上的 vLLM 检查点比较。","Nemotron-3-Embed-8B-BF16 通过将 Ministral-3-8B-Instruct-2512：https://huggingface.co/mistralai/Ministral-3-8B-Instruct-2512 的因果解码器转换为用于全序列检索的双向编码器，从而适配其主干结构。该模型使用网络来源和合成文本对的混合进行对比预训练，然后在精心策划的多语言检索数据集上进行微调，涵盖法律、金融、医疗、商业和教育等领域。该 8B 模型作为旗舰嵌入模型，同时来自同一开发线的早期 8B 教师检查点被用来蒸馏高效的 1B 变体。","1B 模型不是从零开始训练的小型检索器。我们首先将双向适配方案应用于 Ministral-3-3B-Instruct-2512：https://huggingface.co/mistralai/Ministral-3-3B-Instruct-2512 主干网络，以建立 3B 检索器基础，然后通过两轮结构化剪枝和蒸馏进行压缩。","首先，使用 NVIDIA ModelOpt：https://github.com/NVIDIA/Model-Optimizer 的 mcore_minitron 神经架构搜索引擎，将 3B 父模型压缩到 2B 中间版体积。NAS 管线在严格的参数预算下搜索隐藏层宽度、FFN 大小、注意力头数和深度，以识别适用于检索工作负载的高效架构。","得到的 2B 中间模型随后通过 8B 教师检查点进行蒸馏，以恢复排序准确性。我们在多语言、同领域的检索数据混合上使用余弦距离损失与均方误差损失的组合，使学生模型的嵌入与教师模型对齐。","：https://cdn-uploads.huggingface.co/production/uploads/697fa5ae089a7f9330c5078f/U6hbjPbLo70KwwtbAs2np.png","图 6. 剪枝和蒸馏管线将检索器从 3B 基础压缩到最终 1B 生产模型。","这一相同流程，即先 ModelOpt 结构化剪枝，然后 8B 教师蒸馏，在第二次重复时，将 2B 中间模型压缩到最终 1.14B 嵌入模型。最终训练使用渐进的两阶段上下文缩放计划：","下表总结了 Nemotron 3 Embed 模型的核心技术规格和部署目标：","表 2. Nemotron 3 Embed 模型的架构规格及核心推理配置。","企业 ISV、原生 AI 公司、数据平台公司和存储提供商已在代理检索、代理记忆、代码检索和生产推理工作流中评估 Nemotron 3 Embed。","NVIDIA Nemotron 3 Embed 以开放权重和开源训练方案发布，使组织能够完全控制检索模型在生产 AI 应用中的定制和部署方式。","开发者可以使用最适合其工作流的部署选项开始使用：","对于需要领域适配或减少占用的工作负载，我们还开源了 NVIDIA NeMo AutoModel：https://github.com/nvidia-nemo/automodel 训练示例：","例如，在 NV Docs：https://huggingface.co/datasets/nvidia/Retrieval-Synthetic-NVDocs-v1 评测中，对 Nemotron-3-Embed-1B-BF16 进行微调，使 NDCG@10 从 56.7% 提升到 63.3%（+11.6%），Recall@5 从 56.1% 提升到 62.8%（+11.9%）。","无论你是在构建企业搜索、生产 RAG、智能体记忆、代码检索，还是自治 AI 系统，NVIDIA Nemotron 3 Embed 都提供灵活的部署选项——从 Hugging Face 上的开源模型，到全托管的 AI 云平台和 NVIDIA NIM 微服务。","探索这些模型，在你偏好的平台上部署，并告诉我们你正在构建什么。","深度研究智能体的公平与解耦评测","恭喜获得 LMEB 的 SOTA！🎉","合并在 https://github.com/embeddings-benchmark/results/pull/617 的结果显示出卓越的 LMEB 性能：https://github.com/embeddings-benchmark/results/pull/617","Nemotron-3-Embed-8B-BF16：64.4，新的整体 SOTA。Nemotron-3-Embed-1B-BF16：61.5，在约 1B 规模上刷新 SOTA。","LMEB 排行榜：https://mteb-leaderboard.hf.space/benchmark/LMEB 似乎正在等待同步。","现在已上线！感谢你在那里评测我们的模型！"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"Aioga 编辑摘要：NVIDIA 发布开源嵌入模型系列 Nemotron 3 Embed，包含三个模型。 Aioga 将其归入「AI资讯」方向，重点关注它对真实使用和行业竞争的影响。","background":"背景分析：模型与研究类动态需要结合能力边界、开放方式、成本、可用性和真实任务表现判断，单项指标领先不等于已经形成稳定采用。","viewpoint":"Aioga 判断：这条动态更适合作为行业观察信号，当前信息足以建立线索，但不足以推导长期结论。","implications":"影响分析：对相关团队而言，短期应先核对来源、可用范围和实际成本，再判断是否值得接入或跟进。","nextStep":"后续观察：继续观察官方文档、实际可用性、价格变化、开发者反馈和竞品回应。","evidenceRefs":["title","summary","articleBody"],"confidence":"medium","status":"published","aiGenerated":false,"autoApproved":true,"generatedBy":"rule-safe-fallback","generatedAt":"2026-07-23T08:10:16.860Z","sourceHash":"b1cd278a5f080393","validation":{"passed":true,"mode":"rule-safe-fallback","checks":["schema","length","source-attribution","no-html"]}},"tags":["AI资讯","Hugging Face：Blog（RSS）"],"translations":{"zh-CN":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源和原文入口，并将继续跟踪官方更新与行业反馈。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga AI资讯","description":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源和原文入口，并将继续跟踪官方更新与行业反馈。","url":"https://www.aioga.com/news/cmrnqu5q801otbizzgvoejz74/"},"en":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI News. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源和原文入口，并将继续跟踪官方更新与行业反馈。","category":"AI News","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga AI News","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI News. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aio","url":"https://www.aioga.com/en/news/cmrnqu5q801otbizzgvoejz74/"},"ja":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aiogaは「AIニュース」の動きとして、Hugging Face：Blog（RSS） からの更新を追跡しています。NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源和原文入口，并将继续跟踪官方更新与行业反馈。","category":"AIニュース","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga AIニュース","description":"Aiogaは「AIニュース」の動きとして、Hugging Face：Blog（RSS） からの更新を追跡しています。NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间","url":"https://www.aioga.com/ja/news/cmrnqu5q801otbizzgvoejz74/"},"ko":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga는 Hugging Face：Blog（RSS）의 업데이트를 AI 뉴스 흐름으로 추적합니다. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源和原文入口，并将继续跟踪官方更新与行业反馈。","category":"AI 뉴스","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga AI 뉴스","description":"Aioga는 Hugging Face：Blog（RSS）의 업데이트를 AI 뉴스 흐름으로 추적합니다. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源","url":"https://www.aioga.com/ko/news/cmrnqu5q801otbizzgvoejz74/"},"es":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga sigue esta actualización de Hugging Face：Blog（RSS） dentro de Noticias IA. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源和原文入口，并将继续跟踪官方更新与行业反馈。","category":"Noticias IA","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga Noticias de IA","description":"Aioga sigue esta actualización de Hugging Face：Blog（RSS） dentro de Noticias IA. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog","url":"https://www.aioga.com/es/news/cmrnqu5q801otbizzgvoejz74/"},"fr":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga suit cette mise à jour de Hugging Face：Blog（RSS） dans la catégorie Actu IA. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Blog（RSS） 发布，Aioga 已同步发布时间、来源和原文入口，并将继续跟踪官方更新与行业反馈。","category":"Actu IA","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga Actualités IA","description":"Aioga suit cette mise à jour de Hugging Face：Blog（RSS） dans la catégorie Actu IA. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该动态由 Hugging Face：Bl","url":"https://www.aioga.com/fr/news/cmrnqu5q801otbizzgvoejz74/"},"de":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga KI-News","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/de/news/cmrnqu5q801otbizzgvoejz74/"},"pt-BR":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga Notícias de IA","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/pt-BR/news/cmrnqu5q801otbizzgvoejz74/"},"ru":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga Новости ИИ","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/ru/news/cmrnqu5q801otbizzgvoejz74/"},"ar":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga أخبار الذكاء الاصطناعي","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/ar/news/cmrnqu5q801otbizzgvoejz74/"},"hi":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga AI समाचार","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/hi/news/cmrnqu5q801otbizzgvoejz74/"},"it":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga Notizie IA","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/it/news/cmrnqu5q801otbizzgvoejz74/"},"nl":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga AI-nieuws","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/nl/news/cmrnqu5q801otbizzgvoejz74/"},"tr":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga AI Haberleri","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/tr/news/cmrnqu5q801otbizzgvoejz74/"},"vi":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Tin tức AI Aioga","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/vi/news/cmrnqu5q801otbizzgvoejz74/"},"id":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Berita AI Aioga","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/id/news/cmrnqu5q801otbizzgvoejz74/"},"th":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - ข่าว AI Aioga","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/th/news/cmrnqu5q801otbizzgvoejz74/"},"pl":{"title":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval","summary":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将继续跟踪后续更新。来源：Hugging Face：Blog（RSS）。","category":"AI资讯","source":"Hugging Face：Blog（RSS）","aggregationSource":"Hugging Face：Blog（RSS）","pageTitle":"NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval - Aioga Wiadomości AI","description":"Aioga tracks this update from Hugging Face：Blog（RSS） under AI资讯. NVIDIA Nemotron 3 Embed Ranks #1 Overall on RTEB， Advancing Agentic Retrieval。该来源目前仅提供简短信息，Aioga 已保留发布时间、来源和原文入口，并将","url":"https://www.aioga.com/pl/news/cmrnqu5q801otbizzgvoejz74/"}}}}